Sensor-based Smart Oven system to enhance cooking safety
Kitchen is the second place where the majority of domestic accidents occur, and in particular oven presents the most principal source of fire accidents in residence. Therefore, enabling kitchen safety is a major factor particularly, for ageing people independent living. The paper presents our sensor-based smart oven system that targets enhancing safety of ageing people while cooking. The system is based on our insightful study on cooking risks analysis and assessment that enables to determine the pertinent parameters to be monitored around oven in order to identify and measure risk situations while cooking. We also introduce in this paper a solution for sensors positioning and system integration in a real-world cooking environment. Furthermore, we present the results of sensors testing in real-world configurations and mainly focus on the results of our experimental study that leads us to select the appropriate sensors that constitute the basic building block of our smart oven system. The system is composed of sensor nodes to monitor events around oven, then the sensory data is transmitted to a computing unit. The system proactively reacts to hazards in order to prevent cooking associated risks.
Results. We identified three major risks during cooking that are: fire, burn or intoxication by gas/smoke. Following are a summary of the studied parameters: 1) Fire: we observed the Volatile Organic Compound (VOC) and Alcohol gases’ concentration in the cooking smoke. 2) Burn: for both, burn risk by splash of a hot liquid and by contacting hot objects, we observed the relative humidity, utensil temperature, burner temperature, and presence of object on burner. 3) Intoxication by gas/smoke: we observed the concentration of CO gas in the cooking smoke.
Smart-oven system overview. Our sensor-based smart oven system allows sensing contextual cooking activities and offering appropriate context-aware interventions. Determining a risk situation and the corresponding interventions are adaptable to user needs.
Our smart oven system is composed of three main modules: A) Contextual data acquisition module. B) Reasoning engine module that determines a risk situation and its severity level based on fuzzy logic. C) Interventions module that is responsible to manage risk situations and to trigger the appropriate interventions.
- Contextual data acquisition module.
The system is based on a smart environment infrastructure, especially sensors and actuators:
- Sensors. Installed around oven to perform context acquisition. They allow the system to infer the situation during cooking, or detect changes in the surrounding environment (e.g., smoke, burner temperature, utensil temperature, and presence of a utensil on burner).
- Actuators. Distributed in the residence to ubiquitously alert user of cooking risk situations. They provide feedback through screens, speakers, or flashing lights, and control appliances in the kitchen (such as switch off oven power). The actuators provide a wide range of possibilities for human-machine interaction including appropriate intervention for each detected risk situation, and an adapted reaction according to user needs.Reasoning engine module.
B. Reasoning engine module.
The reasoning engine is based on our risk detection algorithms for each type of risk: fire, burn by splash, burn by contacting hot objects, and intoxication by CO gas. We present in the paper our threshold-based risk detection algorithms with three risk levels that correspond to two threshold values of the input parameters for each risk. The reasoning engine manages the detection of risk situations and determines their severity levels according to the contextual information around oven. Our reasoning engine is based on fuzzy logic approach that enables to efficiently implement an adaptable and flexible reasoning engine and to improve the compromise between generating false alarms, and an early accurate detection of a risk situation, as well as its severity level management, in presence of multiple input parameters.
We briefly mention our fire risk detection algorithm:
Fire-Risk-Level = 0
If (concentration of VOC or concentration of Alcohol between 200 and 250 ppm) then
Fire-Risk-Level = 1
Else
If (concentration of VOC or concentration of Alcohol > 250 ppm) then
Fire-Risk-Level = 2
End if
C. Interventions engine module.
The intervention engine module triggers the appropriate interventions based on our intervention protocol presented in the paper. The interventions are triggered according to the type of the detected risk and its severity level . For example, if the level of the detected risk is low then, the intervention engine ubiquitously notifies the user about the current detected risk through screens, appropriate lights, and speakers in the rooms of the residence and switches off the oven power. If the level of the risk does not diminish during a predetermined amount of time a second level of interventions will be triggered by the intervention engine. A second level of interventions enables to notify the user, and/or family members and/or caregivers through mobile phone, and messages.
